Airport bus scheduling scheme optimization method and related equipment

By collaboratively optimizing the timetable and bus scheduling of airport buses, and using the improved SPEA2 algorithm, the problems of high operating costs and low passenger satisfaction in the existing technology are solved, achieving more efficient resource allocation and service quality improvement.

CN118966443APending Publication Date: 2024-11-15CHANGAN UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411058510.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-02
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

When the existing airport bus operation planning model faces complex and changing passenger flow dynamics, there is a conflict between schedule optimization and bus scheduling, resulting in an increase in operating costs and a decrease in passenger satisfaction. The fixed vehicle model strategy cannot flexibly cope with different time periods. Changes in passenger flow have led to waste of resources and limited operational benefits.

Method used

Adopting an airport bus scheduling solution optimization method, by obtaining operation time, line information, bus resources and passenger flow requirements, combined with the improved SPEA2 algorithm, coordinate the optimization of timetables and bus scheduling, flexibly configure multiple lines and multiple models to adapt to different types of vehicles. Passenger flow demand during the period, optimize departure timetable and scheduling plans.

Benefits of technology

The coordinated optimization of timetables and bus scheduling has been achieved, which has improved bus utilization, reduced operating costs, reduced passenger waiting time and operating costs, improved service quality, and promoted the development of airport bus systems toward intelligence and efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118966443A_ABST
    Figure CN118966443A_ABST
Patent Text Reader

Abstract

The invention discloses an airport bus scheduling scheme optimization method and related equipment. The method comprises the following steps: acquiring airport bus operation time, airport bus operation line information, bus resource information and passenger flow personnel demand information; and inputting the airport bus operation line information, the bus resource information and the number of passengers arriving in each direction of each operation line of the airport bus in each time period into an airport bus scheduling scheme optimization model, and solving the airport bus scheduling scheme optimization model by combining an improved SPEA2 algorithm. And obtaining a departure time table set in two directions of multiple operation lines of the airport bus of the multi-target balance, and an airport bus scheduling scheme set for executing the departure time table set. According to the method, collaborative optimization of the timetable and bus scheduling can be realized, passenger flow demands in different time periods can be flexibly coped, an airport bus timetable and a bus scheduling scheme which better meet actual demands can be planned, the bus utilization rate is improved, the operation cost is reduced, and thus an airport bus system is promoted to develop towards a more intelligent and efficient direction.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent transportation systems in the field of transportation, and specifically relates to an airport bus dispatching scheme optimization method and related equipment. Background Art

[0002] Airport bus is an important mode of transportation connecting airports and urban areas. With the increasing demand for air transportation, a reasonable airport bus schedule and bus dispatching plan are essential. Existing airport buses usually use traditional operation planning plans, that is, route planning, schedule optimization, bus dispatching and personnel dispatching are carried out in sequence. Schedule optimization and bus dispatching often conflict with each other. A good schedule may increase the difficulty of bus dispatching and increase operating costs; conversely, saving bus dispatching costs may make it impossible to design an excellent schedule, resulting in a decline in the service level of airport buses. In addition, each route of airport buses uses a fixed type of bus, which will lose passengers when the number of passengers increases, and the empty load rate will increase when the number of passengers decreases, affecting passenger satisfaction and operating costs. In the actual operation process, airport buses also have the characteristics of large differences in passenger flow characteristics in the two directions of airport to urban area (upward) and urban area to airport (downward). Therefore, considering the balance between passenger satisfaction and enterprise operating costs, it is very practical to study the schedule formulation and bus dispatching of multiple models on multiple routes at the same time.

[0003] In the modern urban transportation system, airport buses are an important link between the core areas of the city and the airport. Their operating efficiency and service quality directly affect the travel experience of passengers and the overall efficiency of urban transportation. With the rapid development of the aviation industry and the growing demand for passenger travel, the airport bus system is facing unprecedented challenges and opportunities. Traditionally, the operation planning of airport buses often follows a linear and relatively independent process, that is, first determine the route layout, then optimize the timetable, and finally arrange the bus and personnel according to the timetable. Although this traditional planning model meets the basic operational needs to a certain extent, it gradually exposes its limitations in the face of complex and changing passenger flow dynamics and increasing operational efficiency requirements. First, the separation of timetable optimization and bus scheduling leads to potential conflicts between the two. The idealized timetable design may ignore the resource constraints and cost considerations in actual bus scheduling, resulting in increased scheduling difficulty and rising operating costs; while the simple pursuit of minimizing bus scheduling costs may sacrifice the rationality of the timetable, thereby affecting the punctuality of bus services and passenger satisfaction. Secondly, the existing airport bus system generally uses fixed models to serve specific routes, which lacks flexibility. During peak hours, if the bus capacity is insufficient, it will be difficult to meet the surging passenger demand, resulting in passenger loss; during non-peak hours, the bus empty rate may be too high, which not only wastes resources but also increases unnecessary operating costs. This single-model strategy fails to fully utilize the differences in passenger capacity and operating costs of different types of buses, limiting the improvement of operating efficiency. In addition, there are significant differences in passenger flow characteristics in the upstream and downstream directions of airport buses, which is often overlooked in existing operation plans. The upstream direction (airport to city) usually faces greater passenger flow pressure during the flight arrival period, while the downstream direction (city to airport) has a larger passenger flow in the period before the flight departure. This asymmetric passenger flow distribution requires that the operation plan must more carefully consider the changes in passenger flow in different directions and at different times to achieve the optimal allocation of resources. Summary of the invention

[0004] In view of the problems existing in the prior art, the present invention provides an airport bus scheduling optimization method and related equipment, which can achieve coordinated optimization of timetables and bus scheduling, flexibly respond to passenger flow demands in different time periods, plan airport bus timetables and bus scheduling plans that are more in line with actual needs, improve bus utilization, and reduce operating costs, thereby promoting the development of the airport bus system towards a more intelligent and efficient direction.

[0005] In order to solve the above technical problems, the present invention is implemented by the following technical solutions:

[0006] According to a first aspect of the present invention, there is provided a method for optimizing an airport bus scheduling scheme, comprising:

[0007] Obtaining the airport bus operating hours, airport bus operating route information, bus resource information and passenger flow demand information; the airport bus operating route information includes the number of operating routes, the driving distance in both directions of each operating route and the first and last departure times in both directions of each operating route; the bus resource information includes the types of buses that can operate normally, the number of buses of different types, the energy consumption cost and fixed departure cost of buses of different types and the average time-sharing driving time of buses on different operating routes; the passenger flow demand information includes the time when each passenger arrives at the station, the operating route to be taken and the itinerary destination;

[0008] The airport bus operating time is divided into set time periods, and the number of passengers arriving in each direction of each operating route of the airport bus in each time period is obtained by combining the time when each passenger arrives at the station, the operating route to be taken, and the travel destination;

[0009] The airport bus operating route information, the bus resource information and the number of passengers arriving in each direction of each operating route of the airport bus in each time period are input into the airport bus scheduling optimization model, and the airport bus scheduling optimization model is solved in combination with the improved SPEA2 algorithm to obtain a departure timetable set for two directions of multiple operating routes of the airport bus with multi-objective trade-offs, as well as an airport bus scheduling plan set for executing the departure timetable set.

[0010] In a possible implementation of the first aspect, the airport bus scheduling optimization model is specifically:

[0011] min Z1=f w +f l

[0012] min Z2=f e +f s

[0013]

[0014]

[0015] stt z =Tb+hz, z=1,…,N

[0016]

[0017] z1, z2=1,…,N,k=1,…,M,r=1,…,R,z2>z1

[0018]

[0019] z1, z2=1,…,N,k=1,…,M,r=1,…,R,z2>z1

[0020] x ikr (Ta ir +t i )≤y jkr t j +(1-y jkr )B,

[0021] i, j=1,...,N,k=1,...,M,r,=1,...,R,j>i

[0022] y jkr (Tc jr +t j )≤x ikr′ t i +(1-x ikr′ )B,

[0023] i, j=1,…,N,k=1,…M,r,r′=1,…,R,i>j

[0024]

[0025] f zir =0,z,i=1,…,N,r=1,…,R,z≥i

[0026] g zjr =0,z,j=1,…,N,r=1,…,R,z≥j

[0027]

[0028] x ikr ∈{0,1}i=1,…,N+1,k=1,…,M,r=1,…,R

[0029] y jkr ∈{0, 1}, j=1,…,N+1,k=1,…,M,r=1,…,R

[0030] f zir , g zjr ∈Z + ,z,i,j=1,…,N+1,r=1,…,R

[0031] In the formula, f w Waiting time for passengers going up and down; l is the passenger's lost time; Z1 is the total waiting time of the passenger; f e is the energy consumption cost of the bus; f sis the fixed departure cost of the bus; Z2 is the operating cost of the airport bus; N is the number of time periods after the airport bus operating time is divided into set time periods; R is the number of operating routes; f zir represents the number of passengers who arrived at the rth operating line at time z and left at time i in the downlink direction; t i is the time value at the i-th time point; t z is the time value at the zth time point; h is the length of the set time period; g zjr represents the number of passengers who arrived at the rth operating line at time z and left at time j in the upward direction; t j is the time value at the jth time point; Pa zr is the number of passengers arriving at the airport waiting point in the zth time period on the rth operating route; is the number of passengers arriving in each direction of each operating route of the airport bus in each time period; α is the penalty value caused by passenger loss; M is the number of buses that can operate normally; Pc zr is the number of passengers arriving at the downtown waiting point on the rth operating line in the zth period; ikr Indicates that in the downward direction of the rth operating line, if the kth bus is dispatched at the i-th time, it is 1, otherwise it is 0; La r is the down-running distance of the rth operating line; Lc r is the up-travel distance of the rth operating line; L is; θ k is the unit energy consumption cost of the kth bus; y jkr Indicates that in the upward direction of the rth operating line, if the kth bus is dispatched at time , it is 1, otherwise it is 0; σ k is the fixed dispatch cost of the kth bus; Tb is the start time of operation; Ba r Ea is the first departure time of the rth operating line in the down direction; r Bc is the last departure time of the rth operating line in the down direction; r Ec is the first departure time of the rth operating line in the upward direction; r is the last departure time of the rth operating line in the upward direction; a r b is the time period corresponding to the first departure time of the rth operating line in the downlink direction; r The time period corresponding to the last departure time of the rth operating line in the downlink direction; c r The time period corresponding to the first departure time of the rth operating line in the upward direction; d r is the time period corresponding to the last departure time of the rth operating line in the upward direction; Wa is the maximum waiting time for passengers going down; B is an infinite constant; Ta ir Tc is the average travel time required for a bus in the downstream direction to complete the task when it leaves the rth operating line at the i-th time; jris the average travel time required for the bus in the upward direction to complete the task when it leaves the rth operating line at the jth time; Wc is the maximum waiting time for the upward passengers; z ku If the kth bus is a u-type bus, it is 1, otherwise it is 0; C k is the capacity of the kth bus; U is the total number of bus types.

[0032] In a possible implementation manner of the first aspect, the improved SPEA2 algorithm specifically includes:

[0033] The first part is the adaptive crossover probability strategy, which calculates the similarity H of the external archive set of two adjacent generations of SPEA2 by calculating the number of identical individuals between two adjacent generations:

[0034]

[0035] Among them, S is the number of individuals in the external archive set of this generation that are the same as those in the previous generation, S max is the maximum value of S, that is, the number of external archive sets; H∈[0,1], the more similar individuals there are in the external archive before and after iteration, the larger the H value is, and the more likely the algorithm is to converge;

[0036] According to the similarity H of the external archive set, the crossover mutation probability is updated in each generation:

[0037]

[0038] Where P cMin and P cMax are the minimum and maximum crossover probability values, P mMin and P mMax are the minimum and maximum values ​​of the mutation probability value respectively;

[0039] The second part is an external archive set of variable size. t Based on the introduction of an external archive set B of uncertain size t , save the historical optimal individuals of different departure schedules, that is, the individuals with the smallest total waiting time of passengers and the operating cost of the airport bus under the same departure schedule, and update them in each iteration process;

[0040] The third part is the population update strategy, which updates the population P of the SPEA2 algorithm. t Perform crossover mutation to generate a new population and combine it with the original population and external archive set B t Mix to get the mixed population P t `, using the fast non-dominated sorting and crowding calculation of the multi-objective genetic algorithm, the mixed population P t `Select and generate offspring population P t+1 .

[0041] In a possible implementation of the first aspect, the improved SPEA2 algorithm is used to solve the airport bus scheduling optimization model, and the specific process is as follows:

[0042] Step 1: Initialize the population size Popsize, the maximum number of iterations Maxiter, and the minimum crossover probability value P cMin and the maximum value P cMax And the minimum value P of the mutation probability value mMin and the maximum value P mMax ;

[0043] Assume that the evolutionary generation t = 0, randomly generate the initial population P0, construct an empty initial external archive set A0 of Paretosize and an initial external archive set B0 of unfixed size;

[0044] Step 2: According to the population P t The corresponding total waiting time of passengers and the operating cost of the airport bus are calculated using the K-nearest neighbor algorithm. t The fitness value of

[0045] Stpe3: According to population P t The fitness value of Paretosize individuals is selected and stored in the external archive set A t ; In SPEA2 original external archive set A t Based on the introduction of an external archive set B of uncertain size t , save the historical optimal individuals of different departure schedules, that is, the individuals with the smallest total waiting time for passengers and the operating cost of the airport bus under the same departure schedule, and update the external archive set B t ;

[0046] Stpe4: If t=0, set the initial crossover probability value Pc t =Pc Max and the initial mutation probability value Pm t =Pm Max ; If t>0, update Pc according to the adaptive crossover mutation probability strategy t and Pm t ;

[0047] Stpe5: According to the population update strategy, the crossover mutation method is used to generate the population P t The offspring population is then combined with the population P t , External Archive Set A t and external archive set B t Mix and select Popsize individuals from them to generate a new population P t+1 ;

[0048] Step 6: If t = Maxiter, terminate the iteration, otherwise, t = t + 1 and return to Step 2.

[0049] According to a second aspect of the present invention, there is provided an airport bus dispatching scheme optimization device, comprising:

[0050] The acquisition module is used to acquire the airport bus operation time, airport bus operation route information, bus resource information and passenger flow personnel demand information; the airport bus operation route information includes the number of operation routes, the driving distance in both directions of each operation route and the first and last departure time in both directions of each operation route; the bus resource information includes the types of buses that can operate normally, the number of buses of different types, the energy consumption cost and fixed departure cost of buses of different types and the average time-sharing driving time of buses on different operation routes; the passenger flow personnel demand information includes the time when each passenger arrives at the station, the operation route to be taken and the itinerary destination;

[0051] A division module is used to divide the airport bus operation time into set time periods, and obtain the number of passengers arriving in each direction of each operation route of the airport bus in each time period in combination with the time when each passenger arrives at the station, the operation route to be taken and the itinerary destination;

[0052] The optimization module is used to input the airport bus operation route information, the bus resource information and the number of passengers arriving in each direction of each operation route of the airport bus in each time period into the airport bus scheduling optimization model, and solve the airport bus scheduling optimization model in combination with the improved SPEA2 algorithm to obtain a multi-objective trade-off airport bus multiple operation routes The departure schedule set in two directions and the airport bus scheduling plan set for executing the departure schedule set.

[0053] According to a third aspect of the present invention, there is provided a device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method for optimizing an airport bus scheduling scheme when executing the computer program.

[0054] According to a fourth aspect of the present invention, there is provided a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the steps of the method for optimizing an airport bus scheduling scheme.

[0055] Compared with the prior art, the present invention has at least the following beneficial effects:

[0056] The present invention provides an airport bus scheduling optimization method, which combines the two stages of timetable design and bus scheduling in public transportation planning, takes the total waiting time of passengers and the operating cost of airport buses as the target, obtains a multi-objective weighed departure timetable set for two directions of multiple operating routes of airport buses, and an airport bus scheduling solution set for executing the departure timetable set, thereby reducing the impact of the conflict between the two stages of timetable design and bus scheduling on airport bus scheduling. The method simultaneously performs airport bus scheduling for two directions of multiple operating routes of airport buses, not only utilizing the different peak hours of bus passenger flow on multiple operating routes, making up for the difference in passenger flow in two directions of airport buses, reducing a large number of empty trips caused by the difference in passenger flow in two directions of airport buses in the past, and reducing the operating cost of airport buses; and also improves the utilization rate of airport buses by sharing airport buses by multiple routes, further reducing the operating cost of airport buses. In addition, the present invention also uses multiple types of airport buses, which can adapt to the changes in airport bus passenger flow, uses small buses when the passenger flow is small, and uses large buses when the passenger flow is large, which is more in line with the actual situation, can effectively reduce the operating cost of airport buses, and can also reduce the waiting time of passengers and improve service quality.

[0057] In summary, the present invention can achieve coordinated optimization of timetable and bus scheduling, flexibly respond to passenger flow demands in different time periods, plan airport bus timetable and bus scheduling plans that better meet actual needs, improve bus utilization, and reduce operating costs, thereby promoting the development of the airport bus system towards a more intelligent and efficient direction.

[0058] Furthermore, the improved SPEA2 algorithm designs an external archive set of non-fixed size and an adaptive crossover and mutation probability strategy to solve the airport bus scheduling optimization model, thereby improving the solution quality of the model.

[0059] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the technical solutions in the specific implementation modes of the present invention, the drawings required for use in the description of the specific implementation modes will be briefly introduced below. Obviously, the drawings described below are some implementation modes of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0061] Figure 1 It is a flow chart of an airport bus dispatching scheme optimization method of the present invention;

[0062] Figure 2This is a graph of the average speed of airport buses versus time;

[0063] Figure 3 It is the time taken by the airport bus to travel and the time relationship diagram;

[0064] Figure 4 It is a schematic diagram of the adaptive crossover mutation probability strategy;

[0065] Figure 5 It is the improved SPEA2 algorithm flow chart;

[0066] Figure 6 : is a data test result diagram of holidays, working days and rest days in the embodiment;

[0067] Figure 7 is a graph showing convergence results of the hybrid heuristic algorithm in the embodiment;

[0068] Figure 8 is a graph showing the reduction in operating costs when the passenger service levels are close in the embodiment;

[0069] Fig. 9 This is a graph showing the improvement of passenger service level when the operating costs are close in the embodiment;

[0070] Fig.10 1. It is a comparison chart of passenger loss between the optimization scheme and the actual scheme for 10 sets of data in the embodiment;

[0071] Fig.11 1. It is a comparison chart of passenger waiting time between the optimization scheme and the actual scheme for 10 sets of data in the embodiment;

[0072] Fig.12 1. It is a comparison chart of bus departure schedules in the optimization scheme and the actual scheme of 10 sets of data in the embodiment;

[0073] Fig.13 It is a comparison chart of the bus departure schedules of the optimized scheme and the actual scheme for 10 groups of data in the embodiment. DETAILED DESCRIPTION

[0074] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0075] like Figure 1As shown, an embodiment of the present invention provides an airport bus scheduling optimization method, which comprehensively considers the operation time, operation route information, bus resources and passenger flow demand of the airport bus, adopts a model algorithm, and realizes the coordinated optimization of the airport bus departure timetable and bus scheduling, which specifically includes the following steps:

[0076] S1. Obtain airport bus operation time, airport bus operation route information, bus resource information and passenger flow demand information.

[0077] The information about the airport bus operation routes includes the number of operation routes, the driving distance in both directions of each operation route, and the first and last departure times of each operation route in both directions, that is, the driving distance in both directions of each operation route (upward: airport to downtown, downward: downtown to airport) and the first and last departure times. The airport bus has multiple operation routes, and each operation route includes two directions: downtown to airport and airport to downtown.

[0078] It should be noted that since the airport bus has fewer stops on each operating route, each operating route is assumed to have only two stops, the airport and the downtown terminal, and the length of each operating route is set to the distance between the airport and the downtown terminal.

[0079] Information about bus resources includes the types of buses that can operate normally, the number of buses of different types, the energy consumption cost and fixed dispatch cost of buses of different types, and the average travel time of buses on different operating routes. In other words, the average travel time of buses on different operating routes is the time required for buses to complete a certain operating route at different times.

[0080] It should be noted that the energy consumption cost and fixed dispatch cost of each type of bus are calculated based on the purchase cost, fuel consumption and other data of each type of airport bus. According to the historical operation data of buses on different operating routes, the average speed of airport buses at various time points is calculated, such as Figure 2 As shown, the time required for the bus to complete the journey of different routes at different time points is calculated using the integral method according to the mileage of different routes, as shown in Figure 3 As shown. Each route of the airport bus has the same airport station and depot, and only the same depot in the city, but the terminal stations in the city are different and far apart. Therefore, it is assumed that the line change operation of the airport bus can only be completed at the airport station, that is, the bus can only perform the up-trip of an operating route after performing the down-trip of the route, and can perform the down-trip of other routes after performing the up-trip of an operating route.

[0081] Information on passenger demand includes the time each passenger arrives at the station, the operating line they need to take, and the destination of their trip. For example, the time each passenger arrives at the station, the operating line they need to take, and the destination of their trip can be collected through ticketing systems, passenger surveys, and other means.

[0082] About the airport bus operating hours, that is, the start and end times of the bus service.

[0083] S2. Divide the airport bus operating time by set time periods, and combine the time when each passenger arrives at the station, the operating route to be taken, and the travel destination to obtain the number of passengers arriving in each direction of each operating route of the airport bus in each time period.

[0084] Specifically, according to the operation characteristics of the airport bus and the travel patterns of passengers, reasonable time periods are set to divide the operation time. Combined with the arrival time of each passenger, the required operation route and the itinerary destination, the number of passengers arriving in each direction on each operation route in each time period is counted to form detailed demand forecast data.

[0085] For example, the airport bus operating time is divided into 10-minute time periods with an accuracy of 10 minutes, and the number of passengers arriving in each direction of each route of the airport bus in each time period is obtained. Since the passenger flow of airport buses is greatly affected by flights, and the passenger flow of flights is relatively stable within a flight season except for sudden influences such as bad weather, and most of the historical passengers of airport buses purchase tickets after asking at the waiting station. Based on the passenger ticket purchase information of the airport bus in a period of time, including the passenger ticket purchase time, ticket purchase route and direction, the airport bus operating time is divided into 10-minute time periods with an accuracy of 10 minutes, and the number of passengers arriving at the waiting station in each direction of each operating route in each time period is calculated.

[0086] S3. Input the airport bus operating route information, the bus resource information and the number of passengers arriving in each direction of each operating route of the airport bus in each time period into the airport bus scheduling optimization model, and solve the airport bus scheduling optimization model in combination with the improved SPEA2 algorithm to obtain a multi-objective trade-off airport bus multiple operating routes in two directions departure schedule set, and an airport bus scheduling plan set for executing the departure schedule set.

[0087] Specifically, the airport bus operating route information, bus resource information and the number of passengers arriving in each direction of each operating route of the airport bus in each time period (i.e., passenger demand forecast data) are input into the airport bus scheduling optimization model, and the improved SPEA2 algorithm is used for iterative solution to finally obtain a multi-objective trade-off set of airport bus departure timetables for multiple operating routes in two directions and the corresponding airport bus scheduling plan set, that is, a bidirectional, multi-route, and multi-model airport bus scheduling plan set.

[0088] It should be noted that the airport bus scheduling solution set includes: 1) the Pareto solution set of minimum total passenger waiting time - minimum bus operating cost, that is, the solution set; 2) the up and down departure timetables of each operating line in each solution; 3) the bus travel task allocation plan for various types of vehicles in each solution.

[0089] The airport bus dispatching scheme of the present invention includes restrictions on the departure and return of the airport bus to the parking lot. The airport bus has two parking lots, one in the city and one in the airport, for storing the buses. If the first trip of the bus is down, the bus departs from the parking lot at the airport, and if the first trip is up, the bus departs from the parking lot in the city. In order to ensure the use of the airport bus in the next operating period, the present invention sets the same number of buses of various types stored in the two parking lots at the airport and the city before and after the operation.

[0090] In one possible implementation, the airport bus scheduling optimization model is specifically as follows:

[0091] min Z1=f w +f l

[0092] min Z2=f e +f s

[0093]

[0094] stt z =Tb+hz, z=1,…,N

[0095]

[0096]

[0097] z1, z2=1,…,N,k=1,…,M,r=1,…,R,z2>z1

[0098]

[0099] z1, z2=1,…,N,k=1,…,M,r=1,…,R,z2>z1

[0100] xikr (Ta ir +t i )≤y jkr t j +(1-y jkr )B,

[0101] i, j=1,...,N,k=1,...,M,r,=1,...,R,j>i

[0102] y jkr (Tc jr +t j )≤x ikr′ t i +(1-x ikr′ )B,

[0103] i, j=1,…,N,k=1,…M,r,r′=1,…,R,i>j

[0104]

[0105]

[0106] f zir =0,z,i=1,…,N,r=1,…,R,z≥i

[0107] g zjr =0,z,j=1,…,N,r=1,…,R,z≥j

[0108]

[0109] x ikr ∈{0,1}i=1,…,N+1,k=1,…,M,r=1,…,R

[0110] y jkr ∈{0, 1}, j=1,…,N+1,k=1,…,M,r=1,…,R

[0111] f zir , g zjr ∈Z + ,z,i,j=1,…,N+1,r=1,…,R

[0112] In the formula, f w Waiting time for passengers going up and down; l is the passenger's lost time; Z1 is the total waiting time of the passenger; f e is the energy consumption cost of the bus; f s is the fixed departure cost of the bus; Z2 is the operating cost of the airport bus; N is the number of time periods after the airport bus operating time is divided into set time periods; R is the number of operating routes; fzir represents the number of passengers who arrived at the rth operating line at time z and left at time i in the downlink direction; t i is the time value at the i-th time point; t z is the time value at the zth time point; h is the length of the set time period; g zjr represents the number of passengers who arrived at the rth operating line at time z and left at time j in the upward direction; t j is the time value at the jth time point; Pa zr is the number of passengers arriving at the airport waiting point in the zth time period on the rth operating route; is the number of passengers arriving in each direction of each operating route of the airport bus in each time period; α is the penalty value caused by passenger loss; M is the number of buses that can operate normally; Pc zr is the number of passengers arriving at the downtown waiting point on the rth operating line in the zth period; ikr Indicates that in the downward direction of the rth operating line, if the kth bus is dispatched at the i-th time, it is 1, otherwise it is 0; La r is the down-running distance of the rth operating line; Lc r is the up-travel distance of the rth operating line; L is; θ k is the unit energy consumption cost of the kth bus; y jkr In the upward direction of the rth operating line, if the kth bus is dispatched at the jth time, it is 1, otherwise it is 0; σ k is the fixed dispatch cost of the kth bus; Tb is the start time of operation; Ba r Ea is the first departure time of the rth operating line in the down direction; r Bc is the last departure time of the rth operating line in the down direction; r Ec is the first departure time of the rth operating line in the upward direction; r is the last departure time of the rth operating line in the upward direction; a r b is the time period corresponding to the first departure time of the rth operating line in the downlink direction; r The time period corresponding to the last departure time of the rth operating line in the downlink direction; c r The time period corresponding to the first departure time of the rth operating line in the upward direction; d r is the time period corresponding to the last departure time of the rth operating line in the upward direction; Wa is the maximum waiting time for passengers going down; B is an infinite constant; Ta ir Tc is the average travel time required for a bus in the downstream direction to complete the task when it leaves the rth operating line at the i-th time; jr is the average travel time required for the bus in the upward direction to complete the task when it leaves the rth operating line at the jth time; Wc is the maximum waiting time for the upward passengers; z kuIf the kth bus is a u-type bus, it is 1, otherwise it is 0; C k is the capacity of the kth bus; U is the total number of bus types.

[0113] In one possible implementation, the improved SPEA2 algorithm specifically includes:

[0114] Combination Figure 4 As shown in the figure, the first part is the adaptive crossover mutation probability strategy, which calculates the similarity H of the external archive set of two adjacent generations of SPEA2 by calculating the number of identical individuals between two adjacent generations:

[0115]

[0116] Among them, S is the number of individuals in the external archive set of this generation that are the same as those in the previous generation, S max is the maximum value of S, that is, the number of external archive sets; H∈[0,1], the more similar individuals there are in the external archive before and after iteration, the larger the H value is, and the more likely the algorithm is to converge;

[0117] According to the similarity H of the external archive set, the crossover mutation probability is updated in each generation:

[0118]

[0119] Where P cMin and P cMax are the minimum and maximum crossover probability values, P mMin and P mMax are the minimum and maximum values ​​of the mutation probability respectively.

[0120] The second part is an external archive set of variable size. t Based on the introduction of an external archive set B of uncertain size t , save the historical optimal individuals of different departure schedules, that is, the individuals with the smallest total waiting time of passengers and the operating cost of airport buses under the same departure schedule, and update them in each iteration process.

[0121] The third part is the population update strategy, which updates the population P of the SPEA2 algorithm. t Perform crossover mutation to generate a new population and combine it with the original population and external archive set B t Mix to get the mixed population P t `, using the fast non-dominated sorting and crowding calculation of the multi-objective genetic algorithm, the mixed population P t `Select and generate offspring population P t+1 .

[0122] In one possible implementation, combining Figure 5As shown, the specific process of solving the airport bus scheduling optimization model by combining the improved SPEA2 algorithm is as follows:

[0123] Step 1: Initialize the population size Popsize, the maximum number of iterations Maxiter, and the minimum crossover probability value P cMin and the maximum value P cMax And the minimum value P of the mutation probability value mMin and the maximum value P mMax ;

[0124] Assume that the evolutionary generation t = 0, randomly generate the initial population P0, construct an empty initial external archive set A0 of Paretosize and an initial external archive set B0 of unfixed size;

[0125] Step 2: According to the population P t The corresponding total waiting time of passengers and the operating cost of the airport bus are calculated using the K-nearest neighbor algorithm. t The fitness value of

[0126] Stpe3: According to population P t The fitness value of Paretosize individuals is selected and stored in the external archive set A t ; In SPEA2 original external archive set A t Based on the introduction of an external archive set B of uncertain size t , save the historical optimal individuals of different departure schedules, that is, the individuals with the smallest total waiting time for passengers and the operating cost of the airport bus under the same departure schedule, and update the external archive set B t ;

[0127] Stpe4: If t=0, set the initial crossover probability value Pc t =Pc Max and the initial mutation probability value Pm t =Pm Max ; If t>0, update Pc according to the adaptive crossover mutation probability strategy t and Pm t ;

[0128] Stpe5: According to the population update strategy, the crossover mutation method is used to generate the population P t The offspring population is then combined with the population P t , External Archive Set A t and external archive set B t Mix and select Popsize individuals from them to generate a new population P t+1 ;

[0129] Step 6: If t = Maxiter, terminate the iteration, otherwise, t = t + 1 and return to Step 2.

[0130] In order to more clearly understand the technical solution of the present invention, a detailed description is given below in conjunction with a specific implementation case.

[0131] An airport bus operating company has multiple operating routes. The company has a parking lot at the airport and in the city for parking buses. The bus that performs the upward task first departs from the city garage, and the bus that performs the downward task first departs from the airport garage. The airport bus operating company uses two types of buses in total, namely a medium-sized bus with a passenger capacity of 21 and a large bus with a passenger capacity of 45. The company sets a departure schedule according to fixed departure intervals based on experience, and dispatches buses based on the schedule. Schedule formulation and bus dispatch are two important steps in operation planning that affect passenger satisfaction and corporate operating costs. A good schedule plan may generate large operating costs, and vice versa. The company uses manual dispatching methods. For the convenience of dispatching, fixed-type buses are used for each route. Since the departure stations of different routes in the city are far apart, each bus only completes the task on one route during the operation period. In actual operation, the passenger flow of different bus routes and directions is different, which is often manifested as less passenger flow in the upward direction than in the downward direction, and higher passenger flow in the main route than in other routes. Buses on different routes cannot be used interchangeably, which will lead to waste of capacity and increased empty load rate. In addition, due to the peak period of flight take-off and landing and the traffic conditions in the urban area, the passenger flow at different times is different during the operation period. The use of fixed-type buses will also lead to waste of capacity. For example, on routes using large buses, when the passenger flow decreases at noon or in the evening, the use of large buses will cause a decrease in the bus full load rate, which will increase the operating costs of the enterprise. On routes using small buses, during the peak hours of the day, small buses cannot meet the needs of passengers, resulting in increased waiting time for passengers and a decline in service levels.

[0132] Specifically, the airport bus operating company currently dispatches airport buses for two parking lots, one at the airport and the other in the city, three routes, two directions, and two types of buses as shown in Table 1.

[0133] Table 1 Data details

[0134]

[0135] According to the actual situation of the above-mentioned airport bus operating company, the airport bus scheduling scheme optimization method provided by the present invention is used for optimization.

[0136] For the 10-day passenger demand information of the airport bus operating company, the optimization scheme of the airport bus scheduling optimization method proposed in the present invention is compared with the actual scheme. In the actual comparison, the departure schedule published by the operating company is used in the uplink direction, and the actual departure data is used in the downlink direction. In the original operation plan, the number of large buses used exceeds 30, and the number of small buses used exceeds 10; in the optimization plan, 15 small buses and 20 large buses are used, respectively. When the passenger service level is close, the optimization results are shown in Tables 5 and 6. Figure 8 As shown in Table 6, the operating cost is significantly reduced when the passenger service level is close. When the enterprise operating costs are close, the optimization results are shown in Table 6 and Fig. 9 As shown, the passenger service level is significantly improved when the operating costs are close.

[0137] Table 5 Comparison results between optimization scheme and actual situation

[0138]

[0139] Table 6 Comparison results between optimization scheme and actual situation

[0140]

[0141]

[0142] from Figure 8 It can be seen that when 10 sets of real passenger flow data are used to optimize the airport bus scheduling scheme optimization method of the present invention, when the objective function 1 is close, that is, the passenger service level is close, the optimized scheme reduces the enterprise operating cost by an average of 9.89% compared with the actual scheme. Fig. 9 It can be seen that when the operating costs of enterprises are similar, the passenger service level of the first and second groups of data is higher, 16.90% and 17.03% respectively, and the passenger service level of the remaining groups of data is relatively close, with an average of 5.58%. According to the data source, the first two groups of data are passenger flow data for the May Day holiday. The original scheduling plan is not conducive to dealing with the situation of large changes in passenger flow during holidays, while the airport bus scheduling plan optimization method proposed in the present invention is still effective in the case of sudden changes in passenger flow.

[0143] Based on the above 10 sets of data, the actual plan and the optimized plan are compared and analyzed in terms of average waiting time per passenger, number of lost passengers, and departure frequency of different types of vehicles. Fig.10 and Fig.11It can be seen that both optimization schemes reduce the number of lost passengers of the airport bus. In terms of average waiting time per capita, the average waiting time per capita of Optimization Scheme 1, i.e., the optimization scheme with similar total waiting time of passengers, is slightly higher than the actual scheme. The average waiting time per capita of Optimization Scheme 2, i.e., the optimization scheme with similar operating cost of the airport bus, is still lower than the actual scheme. Fig.12 and Fig.13 It can be seen that both Optimization Scheme 1 and Optimization Scheme 2 reduce the number of large airport bus services and increase the number of small airport bus services, which is more in line with the actual situation where the airport bus share rate is low and the passenger flow is small.

[0144] In order to verify the effectiveness of the improved SPEA2 algorithm of the present invention, the above-mentioned airport bus operating company data and the airport bus scheduling optimization model are used for calculation, and the solution results of the multi-objective genetic algorithm, the K-means improved multi-objective genetic algorithm and the SPEA2 algorithm are compared. The two SP indicators and the HV indicator are also used for comparison.

[0145] like Figure 6 The figure shows the comparison results of the four algorithms under three different test data. It can be found that although the SPEA2 algorithm can converge to the Pareto frontier, the convergence range is small; while the solution space of the NSGA2 algorithm and the improved NSGA2 algorithm has increased, but the solution quality is inferior to the improved SPEA2 algorithm. The Pareto solution obtained by the improved SPEA2 algorithm is closer to the coordinate axis, and the Pareto solution set obtained is more uniform and converges within the required range. Therefore, the improvement of the SPEA2 algorithm is effective.

[0146] In order to further verify the performance of the improved SPEA2 algorithm, the improved SPEA2 algorithm is compared with the comparison algorithm in terms of performance indicators. For each algorithm, it is run independently 10 times to obtain the mean (Mean) and standard deviation (Std) results of the SP indicator. Table 2 shows the comparison results of the SP indicator.

[0147] Table 2 Comparison of SP indicators between the improved SPEA2 algorithm and three comparison algorithms

[0148]

[0149] The improved SPEA2 algorithm and three comparison algorithms are used to solve three sets of test data. For the convenience of comparison, for each test function, the average results of the better SP indicators in Table 2 are bolded. It can be seen that the improved SPEA2 algorithm can obtain better SP indicator mean values ​​compared with other algorithms, and has obvious advantages over other algorithms. Therefore, the Pareto solution set obtained by the improved SPEA2 algorithm has better diversity. In addition, from the standard deviation value of the SP indicator, the standard deviation value of the improved SPEA2 algorithm is the smallest in all 10 sets of test data, and the improved SPEA2 algorithm is at a relatively stable level. Therefore, it can be shown that the improved SPEA2 algorithm has a more stable optimization effect.

[0150] Table 3 shows the mean (Mean) and standard deviation (Std) results of the HV index obtained when the improved SPEA2 algorithm and the other three comparison algorithms solve 10 sets of test data. For the convenience of comparison, for each set of test data, the mean results of the better HV indexes in Table 5 are bolded.

[0151] Table 3 Comparison of HV indicators of the improved SPEA2 algorithm and three comparison algorithms

[0152]

[0153] The HV index in Table 3 reflects the comprehensive performance of the algorithm in terms of convergence and diversity. It can be seen that the improved SPEA2 algorithm can obtain a better HV index mean value than other algorithms, and is higher than the other three algorithms. From the standard deviation of the HV index, the standard deviation of the ISPEA2 algorithm is higher than that of the SPEA2 algorithm in the three test data sets, and the rest are all the best. Figure 7 As shown in the figure, the SPEA2 algorithm is prone to fall into the local optimal solution during the iteration process, and the multiple solution results are close, which leads to a small standard deviation of the HV index of the SPEA2 algorithm. Therefore, considering the convergence of the algorithm comprehensively, it can be shown that the comprehensive performance of the ISPEA2 algorithm is better than other algorithms.

[0154] Table 4 shows the average solution time of the four algorithms on different test data sets. It can be seen from the table that the calculation time of the ISPEA2 algorithm will increase after the improvement, which is 55.177s, 51.1026s, and 42.1543s longer than the other three algorithms. The time period calculated by the test data of this embodiment is 20h. The calculation time increased by the ISPEA2 algorithm is very small compared to the calculated time period, so the time increased by the improved SPEA2 algorithm can be ignored. Comprehensive analysis of the above solution results, SP index, HV index and solution time shows that the improved SPEA2 algorithm can obtain good optimization effects when solving the TT-VS problem proposed in this embodiment, and has certain advantages.

[0155] Table 4 Comparison of solution time of the improved SPEA2 algorithm and three comparison algorithms

[0156]

[0157] Based on the above analysis, the improved SPEA2 algorithm can achieve good optimization results, has certain advantages, and can effectively solve the optimization model of the airport bus scheduling scheme.

[0158] In one embodiment of the present invention, a device for optimizing an airport bus scheduling scheme is provided, comprising:

[0159] The acquisition module is used to obtain the airport bus operation time, airport bus operation route information, bus resource information and passenger flow demand information; the airport bus operation route information includes the number of operation routes, the driving distance in both directions of each operation route and the first and last departure times in both directions of each operation route; the bus resource information includes the types of buses that can operate normally, the number of different types of buses, the energy consumption cost and fixed departure cost of different types of buses and the average time-sharing driving time of buses on different operation routes; the passenger flow demand information includes the time when each passenger arrives at the station, the operation route to be taken and the itinerary destination.

[0160] The division module is used to divide the airport bus operating time into set time periods, and combine the time when each passenger arrives at the station, the operating route to be taken and the travel destination to obtain the number of passengers arriving in each direction of each operating route of the airport bus in each time period.

[0161] The optimization module is used to input the airport bus operation route information, the bus resource information and the number of passengers arriving in each direction of each operation route of the airport bus in each time period into the airport bus scheduling optimization model, and solve the airport bus scheduling optimization model in combination with the improved SPEA2 algorithm to obtain a multi-objective trade-off airport bus multiple operation routes The departure schedule set in two directions and the airport bus scheduling plan set for executing the departure schedule set.

[0162] All relevant contents of each step involved in the aforementioned embodiment of an airport bus scheduling scheme optimization method can be referred to the functional description of the functional module corresponding to an airport bus scheduling scheme optimization device in the embodiment of the present invention, and will not be repeated here. The division of modules in the embodiment of the present invention is schematic, which is only a logical function division. There may be other division methods in actual implementation. In addition, each functional module in each embodiment of the present invention can be integrated into a processor, or it can exist physically separately, or two or more modules can be integrated into one module. The above-mentioned integrated module can be implemented in the form of hardware or in the form of software functional modules.

[0163] In another embodiment of the present invention, a computer device is provided, the computer device comprising a processor and a memory, the memory being used to store a computer program, the computer program comprising program instructions, and the processor being used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or may be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc., which are the computing core and control core of the terminal, which are suitable for implementing one or more instructions, and are specifically suitable for loading and executing one or more instructions in a computer storage medium to implement a corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used for the operation of an airport bus scheduling optimization method.

[0164] In another embodiment of the present invention, the present invention also provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device for storing programs and data. It is understandable that the computer-readable storage medium here can include both built-in storage media in the computer device and, of course, extended storage media supported by the computer device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the above-mentioned embodiment of a method for optimizing an airport bus scheduling scheme.

[0165] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0166] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0167] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0168] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0169] In the present invention, the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" etc. mean that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the described specific features, structures, materials or characteristics may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0170] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for optimizing airport bus scheduling, characterized in that: include: Obtain airport bus operation time, airport bus operation route information, bus resource information and passenger flow demand information; The airport bus operation route information includes the number of operation routes, the driving distance in both directions of each operation route, and the first and last departure times in both directions of each operation route; the bus resource information includes the types of buses that can operate normally, the number of buses of different types, the energy consumption cost and fixed departure cost of buses of different types, and the average travel time of buses in different operation routes; The passenger flow demand information includes the time when each passenger arrives at the station, the operating line to be taken and the itinerary destination; The airport bus operating time is divided into set time periods, and the number of passengers arriving in each direction of each operating route of the airport bus in each time period is obtained by combining the time when each passenger arrives at the station, the operating route to be taken, and the travel destination; The airport bus operating route information, the bus resource information and the number of passengers arriving in each direction of each operating route of the airport bus in each time period are input into the airport bus scheduling optimization model, and the airport bus scheduling optimization model is solved in combination with the improved SPEA2 algorithm to obtain a departure timetable set for two directions of multiple operating routes of the airport bus with multi-objective trade-offs, as well as an airport bus scheduling plan set for executing the departure timetable set.

2. The method for optimizing airport bus scheduling according to claim 1, characterized in that: The airport bus scheduling optimization model is specifically as follows: min Z1=f w +f l min Z2=f e +f s s.t.t z =Tb+hz,z=1,…,N z1, z2=1,…,,k=1,…,M,r=1,…,R,z2>z1 z1, z2=1,…,N,k=1,…,,r=1,…,R,z2>z1 x ikr (Are ir +t i )≤y jkr t j +(1-y jkr )B, i, j=1,...,N,k=1,...,M,r,=1,...,R,j>i y jkr (Tc jr +t j )≤x ikr′ t i +(1-x ikr′ )B, i, j=1,…,N,k=1,…M,r,r′=1,…,R,i>j f zir =0,z,i=1,…,N,r=1,…,R,z≥i g zjr =0,z,j=1,…,N,r=1,…,R,z≥j x ikr ∈{0,1}i=1,…,N+1,k=1,…,M,r=1,…,R y jkr ∈{0,1},j=1,…,N+1,k=1,…,M,r=1,…,R f zir ,g zjr ∈Z + ,z,i,j=1,…,N+1,r=1,…,R In the formula, f w Waiting time for passengers going up and down; l is the passenger's lost time; Z1 is the total waiting time of the passenger; f e is the energy consumption cost of the bus; f s is the fixed departure cost of the bus; Z2 is the operating cost of the airport bus; N is the number of time periods after the airport bus operating time is divided into set time periods; R is the number of operating routes; f zir represents the number of passengers who arrived at the rth operating line at time z and left at time i in the downlink direction; t i is the time value at the i-th time point; t z is the time value at the zth time point; h is the length of the set time period; g zjr represents the number of passengers who arrived at time z on the rth operating line in the upward direction and left at time t; j is the time value at the jth time point; Pa zr is the number of passengers arriving at the airport waiting point on the rth operating route in the zth time period; is the number of passengers arriving in each direction of each operating route of the airport bus in each time period; α is the penalty value caused by passenger loss; M is the number of buses that can operate normally; Pc zr is the number of passengers arriving at the downtown waiting point on the rth operating line in the zth period; ikr Indicates that in the downward direction of the rth operating line, if the kth bus is dispatched at the i-th time, it is 1, otherwise it is 0; La r is the down-running distance of the rth operating line; Lc r is the up-travel distance of the rth operating line; L is; θ k is the unit energy consumption cost of the kth bus; y jkr In the upward direction of the rth operating line, if the kth bus is dispatched at the jth time, it is 1, otherwise it is 0; σ k is the fixed dispatch cost of the kth bus; Tb is the start time of operation; Ba r Ea is the first departure time of the rth operating line in the down direction; r Bc is the last departure time of the rth operating line in the down direction; r The first departure time of the rth operating line in the upward direction; Ec r is the last departure time of the rth operating line in the upward direction; a r b is the time period corresponding to the first departure time of the rth operating line in the downlink direction; r The time period corresponding to the last departure time of the rth operating line in the downlink direction; c r The time period corresponding to the first departure time of the rth operating line in the upward direction; d r is the time period corresponding to the last departure time of the rth operating line in the upward direction; Wa is the maximum waiting time for passengers going down; B is an infinite constant; Ta ir is the average travel time required for a bus in the downstream direction to complete the task when it leaves the rth operating line at the i-th time; Tc jr is the average travel time required for a bus in the upward direction to complete the task when it leaves the rth operating route at the jth time; Wc is the maximum waiting time for upward passengers; ku If the kth bus is a u-type bus, it is 1, otherwise it is 0; C k is the capacity of the kth bus; U is the total number of bus types.

3. The method for optimizing airport bus scheduling according to claim 2, characterized in that: The improved SPEA2 algorithm specifically includes: The first part is the adaptive crossover probability strategy, which calculates the similarity H of the external archive set of two adjacent generations of SPEA2 by calculating the number of identical individuals between two adjacent generations: Among them, S is the number of individuals in the external archive set of this generation that are the same as those in the previous generation, S max is the maximum value of S, that is, the number of external archive sets; H∈[0,1], the more similar individuals there are in the external archive before and after iteration, the larger the H value is, and the more likely the algorithm is to converge; According to the similarity H of the external archive set, the crossover mutation probability is updated in each generation: Where P cMin and P cMax are the minimum and maximum crossover probability values, P mMin and P mMax are the minimum and maximum values ​​of the mutation probability value respectively; The second part is an external archive set of variable size. t Based on the introduction of an external archive set B of uncertain size t , save the historical optimal individuals of different departure schedules, that is, the individuals with the smallest total waiting time of passengers and the operating cost of the airport bus under the same departure schedule, and update them in each iteration process; The third part is the population update strategy, which updates the population P of the SPEA2 algorithm. t Perform crossover mutation to generate a new population and combine it with the original population and external archive set B t Mix to get the mixed population P t `, using the fast non-dominated sorting and crowding calculation of the multi-objective genetic algorithm, the mixed population P t `Select and generate offspring population P t+1 .

4. The method for optimizing airport bus scheduling according to claim 3, characterized in that: The improved SPEA2 algorithm is used to solve the airport bus scheduling optimization model. The specific process is as follows: Step 1: Initialize the population size Popsize, the maximum number of iterations Maxiter, and the minimum crossover probability value P cMin and the maximum value P cMax And the minimum value P of the mutation probability value mMin and the maximum value P mMax ; Assume that the evolutionary generation t = 0, randomly generate the initial population P0, construct an empty initial external archive set A0 of Paretosize and an initial external archive set B0 of unfixed size; Step 2: According to the population P t The corresponding total waiting time of passengers and the operating cost of the airport bus are calculated using the K-nearest neighbor algorithm. t The fitness value of Stpe3: According to population P t The fitness value of Paretosize individuals is selected and stored in the external archive set A t ; In SPEA2 original external archive set A t Based on the introduction of an external archive set B of uncertain size t , save the historical optimal individuals of different departure schedules, that is, the individuals with the smallest total waiting time for passengers and the operating cost of the airport bus under the same departure schedule, and update the external archive set B t ; Stpe4: If t=0, set the initial crossover probability value Pc t =Pc Max and the initial mutation probability value Pm t =Pm Max ; If t>0, update Pc according to the adaptive crossover mutation probability strategy t and Pm t ; Stpe5: According to the population update strategy, the crossover mutation method is used to generate the population P t The offspring population is then combined with the population P t , External Archive Set A t and external archive set B t Mix and select Popsize individuals from them to generate a new population P t+1 ; Step 6: If t = Maxiter, terminate the iteration, otherwise, t = t + 1 and return to Step 2.

5. An airport bus dispatching scheme optimization device, characterized in that: include: The acquisition module is used to obtain the airport bus operation time, airport bus operation route information, bus resource information and passenger flow personnel demand information; The airport bus operation route information includes the number of operation routes, the driving distance in both directions of each operation route, and the first and last departure times in both directions of each operation route; the bus resource information includes the types of buses that can operate normally, the number of buses of different types, the energy consumption cost and fixed departure cost of buses of different types, and the average travel time of buses in different operation routes; The passenger flow demand information includes the time when each passenger arrives at the station, the operating line to be taken and the itinerary destination; A division module is used to divide the airport bus operation time into set time periods, and obtain the number of passengers arriving in each direction of each operation route of the airport bus in each time period in combination with the time when each passenger arrives at the station, the operation route to be taken and the itinerary destination; The optimization module is used to input the airport bus operation route information, the bus resource information and the number of passengers arriving in each direction of each operation route of the airport bus in each time period into the airport bus scheduling optimization model, and solve the airport bus scheduling optimization model in combination with the improved SPEA2 algorithm to obtain a multi-objective trade-off airport bus multiple operation routes The departure schedule set in two directions and the airport bus scheduling plan set for executing the departure schedule set.

6. The device for optimizing airport bus dispatching scheme according to claim 5, characterized in that: The airport bus scheduling optimization model is specifically as follows: min Z1=f w +f l min Z2=f e +f s s.t.t z =Tb+hz,z=1,…,N z1, z2=1,…,N,k=1,…,M,r=1,…,R,z2>z1 z1, z2=1,…,N,k=1,…,M,r=1,…,R,z2>z1 x ikr (Are ir +t i )≤y jkr t j +(1-y jkr )B, i, j=1,...,N,k=1,...,M,r,=1,...,R,j>i y jkr (Tc jr +t j )≤x ikr′ t i +(1-x ikr′ )B, i, j=1,…,N,k=1,…M,r,r′=1,…,R,i>j f zir =0,z,i=1,…,N,r=1,…,R,z≥i g zjr =0,z,j=1,…,N,r=1,…,R,z≥j x ikr ∈{0,1}i=1,…,N+1,k=1,…,M,r=1,…,R y jkr ∈{0,1},j=1,…,N+1,k=1,…,M,r=1,…,R f zir ,g zjr ∈Z + ,z,i,j=1,…,N+1,r=1,…,R In the formula, f w Waiting time for passengers going up and down; l is the passenger's lost time; Z1 is the total waiting time of the passenger; f e is the energy consumption cost of the bus; f s is the fixed departure cost of the bus; Z2 is the operating cost of the airport bus; N is the number of time periods after the airport bus operating time is divided into set time periods; R is the number of operating routes; f zir represents the number of passengers who arrived at the rth operating line at time z and left at time i in the downlink direction; t i is the time value at the i-th time point; t z is the time value at the zth time point; h is the length of the set time period; g zjr represents the number of passengers who arrived at the rth operating line at time z and left at time j in the upward direction; t j is the time value at the jth time point; Pa zr is the number of passengers arriving at the airport waiting point on the rth operating route in the zth time period; is the number of passengers arriving in each direction of each operating route of the airport bus in each time period; α is the penalty value caused by passenger loss; M is the number of buses that can operate normally; Pc zr is the number of passengers arriving at the downtown waiting point on the rth operating line in the zth period; ikr Indicates that in the downward direction of the rth operating line, if the kth bus is dispatched at the i-th time, it is 1, otherwise it is 0; La r is the down-running distance of the rth operating line; Lc r is the up-travel distance of the rth operating line; L is; θ k is the unit energy consumption cost of the kth bus; y jkr In the upward direction of the rth operating line, if the kth bus is dispatched at the jth time, it is 1, otherwise it is 0; σ k is the fixed dispatch cost of the kth bus; Tb is the start time of operation; Ba r Ea is the first departure time of the rth operating line in the down direction; r Bc is the last departure time of the rth operating line in the down direction; r The first departure time of the rth operating line in the upward direction; Ec r is the last departure time of the rth operating line in the upward direction; a r b is the time period corresponding to the first departure time of the rth operating line in the downlink direction; r The time period corresponding to the last departure time of the rth operating line in the downlink direction; c r The time period corresponding to the first departure time of the rth operating line in the upward direction; d r is the time period corresponding to the last departure time of the rth operating line in the upward direction; Wa is the maximum waiting time for passengers going down; B is an infinite constant; Ta ir is the average travel time required for a bus in the downstream direction to complete the task when it leaves the rth operating line at the i-th time; Tc jr is the average travel time required for a bus in the upward direction to complete the task when it leaves the rth operating route at the jth time; Wc is the maximum waiting time for upward passengers; ku If the kth bus is a u-type bus, it is 1, otherwise it is 0; C k is the capacity of the kth bus; U is the total number of bus types.

7. The device for optimizing airport bus dispatching scheme according to claim 6, characterized in that: The improved SPEA2 algorithm specifically includes: The first part is the adaptive crossover probability strategy, which calculates the similarity H of the external archive set of two adjacent generations of SPEA2 by calculating the number of identical individuals between two adjacent generations: Among them, S is the number of individuals in the external archive set of this generation that are the same as those in the previous generation, S max is the maximum value of S, that is, the number of external archive sets; H∈[0,1], the more similar individuals there are in the external archive before and after iteration, the larger the H value is, and the more likely the algorithm is to converge; According to the similarity H of the external archive set, the crossover mutation probability is updated in each generation: Where P cMin and P cMax are the minimum and maximum crossover probability values, P mMin and P mMax are the minimum and maximum values ​​of the mutation probability value respectively; The second part is an external archive set of variable size. t Based on the introduction of an external archive set B of uncertain size t , save the historical optimal individuals of different departure schedules, that is, the individuals with the smallest total waiting time of passengers and the operating cost of the airport bus under the same departure schedule, and update them in each iteration process; The third part is the population update strategy, which updates the population P of the SPEA2 algorithm. t Perform crossover mutation to generate a new population and combine it with the original population and external archive set B t Mix to get the mixed population P t `, using the fast non-dominated sorting and crowding calculation of the multi-objective genetic algorithm, the mixed population P t `Select and generate offspring population P t+1 .

8. The device for optimizing airport bus dispatching scheme according to claim 7, characterized in that: The improved SPEA2 algorithm is used to solve the airport bus scheduling optimization model. The specific process is as follows: Step 1: Initialize the population size Popsize, the maximum number of iterations Maxiter, and the minimum crossover probability value P cMin and the maximum value P cMax And the minimum value P of the mutation probability value mMin and the maximum value P mMax ; Assume that the evolutionary generation t = 0, randomly generate the initial population P0, construct an empty initial external archive set A0 of Paretosize and an initial external archive set B0 of unfixed size; Step 2: According to the population P t The corresponding total waiting time of passengers and the operating cost of the airport bus are calculated using the K-nearest neighbor algorithm. t The fitness value of Stpe3: According to population P t The fitness value of Paretosize individuals is selected and stored in the external archive set A t ; In SPEA2 original external archive set A t Based on the introduction of an external archive set B of uncertain size t , save the historical optimal individuals of different departure schedules, that is, the individuals with the smallest total waiting time for passengers and the operating cost of the airport bus under the same departure schedule, and update the external archive set B t ; Stpe4: If t=0, set the initial crossover probability value Pc t =Pc Max and the initial mutation probability value Pm t =Pm Max ; If t>0, update Pc according to the adaptive crossover mutation probability strategy t and Pm t ; Stpe5: According to the population update strategy, the crossover mutation method is used to generate the population P t The offspring population is then combined with the population P t , External Archive Set A t and external archive set B t Mix and select Popsize individuals from them to generate a new population P t+1 ; Step 6: If t = Maxiter, terminate the iteration, otherwise, t = t + 1 and return to Step 2.

9. A device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the airport bus scheduling optimization method as described in any one of claims 1 to 4 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the airport bus scheduling optimization method as described in any one of claims 1 to 4 are implemented.